In multiple sclerosis (MS), breakdown of the blood-brain barrier (BBB) occurs early in lesion formation and correlates with acute clinical exacerbation. Patients with contrast enhancing plaques are more likely to have irreversible injury, as marked by spectroscopy and T1 hypointensities. BBB disruption leads to edema, and allows CNS entry of inflammatory mediators that exacerbate neuropathology and restrict the capacity for repair. By limiting entrance of pathogenic cells and inflammatory factors, permanent damage can be reduced. New therapies directed at BBB breakdown are needed, as presently only high dose corticosteroids effectively treat exacerbations, and their mechanism of action at the BBB is poorly characterized. Understanding BBB disruption may lead to new therapies to restrict acute exacerbation in MS. The BBB exists at the level of brain microvessel endothelial cells (BMVEC), which use tight junctions to restrict paracellular permeability. Transmembrane proteins contributing to tight junction strands include claudins (CLN) and occludin (OCLN). In BMVEC, CLN-5 plays a key role in determining barrier properties, and CLN5-/- mice display BBB disruption and die perinatally. OCLN regulates properties of the BBB, but is not required for barrier formation. Interestingly, establishment of the BBB is not intrinsic to BMVEC, and depends on astrocytes and pericytes. These cells are also strongly implicated in BBB disruption, but the underlying mechanisms are not well understood. We have identified a novel link between astrocyte reactivity and BBB breakdown. Using microarray analysis of human astrocytes, we have found that cytokines expressed in MS induce genes that cause endothelial plasticity. Upregulated transcripts include the angiogenic factor VEGF-A and its transcriptional regulator, HIF-1a. VEGF-A is a potent inducer of BBB disruption and localizes to reactive astrocytes in MS lesions, but the mechanism underlying its effects on the BBB is unknown. Our studies now reveal that VEGF-A disrupts expression of both CLN-5 and OCLN in BMVEC in vitro, and in the CNS. Downregulation of both proteins accompanies induction of VEGF-A and BBB breakdown in EAE, a widely- used animal model of MS. Our rescue experiments in vitro implicate loss of CLN-5 as a key event in BBB opening. Importantly, we have now generated conditional knockout GfapCre:Vegffl/fl and GfapCre:Hif1afl/fl mice to define the roles of VEGF-A and HIF-1a in BBB breakdown. Our data show that BBB disruption is strikingly restricted in both genotypes. Here, we will test the hypothesis that activation of the HIF-VEGF axis in reactive astrocytes promotes BBB breakdown via disruption of endothelial CLN-5 and OCLN. In Aim#1, we will define the mechanism underlying VEGF-induced permeability of CNS microvascular endothelium. In Aim#2, we will use GfapCre:Vegffl/fl and GfapCre:Hif1afl/fl mice to confirm the roles of HIF-1a and VEGF-A in CLN-5 and OCLN disruption and BBB breakdown in vivo. We will also examine whether rescue of endothelial CLN-5 or OCLN restricts BBB disruption. In Aim#3, using GfapCre:Vegffl/fl and GfapCre:Hif11fl/fl mice we will address the relevance of these findings to disease expression in MOG35-55 EAE. This proposal aims to identify pathways responsible for BBB disruption in MS. The long-term goal of this work is to develop new therapies to restrict lesion formation and limit the severity of acute exacerbation in MS patients.